Cian O'Donnell

Reminder everyone that #SNUFA is next week! Make sure to register in advance to get emailed the link to join.

We’re fighting for Donegal tweed Inside the weavers’ campaign to win protected status for the Irish luxury textile

It's only Donegal tweed if it's from the Tír Chonaill region of Ireland. Otherwise it's just sparkling wool 😉

Good article on the fight to get Donegal tweed a protected status www.ft.com/content/54f5...

Molecular mechanisms mediating engram ensemble retrievability state in mice Engrams, ensembles of neurons that store memories, exist along a continuum of retrievability. Normally, sensory cues can reactivate a latent engram to…

Excited to share the latest from the lab on the molecular mechanisms of silencing and un-silencing a hippocampal engram ensemble. Was fun using a variant of the EPSILON technique developed by Adam Cohen's lab.
(1/2)

www.sciencedirect.com/science/arti...

Sensory expectations shape neural population dynamics in motor circuits Nature - Experiments with human volunteers and macaques show that expectations produced by probabilistic cueing of future sensory inputs shape motor circuit dynamics in order to increase the...

Thrilled that our paper is out today in Nature!
www.nature.com/articles/s4...

I enjoy Janelia's approach, they have a large side screen visible to everyone in the room which has a countdown timer from eg 20 mins to zero during allotted speaker time slot. When it hits zero it turns red and starts counting up again... permits no excuses 😅

Falling asleep follows a predictable bifurcation dynamic - Nature Neuroscience Li et al. propose a conceptual framework to study the phenomenon of falling asleep based on electroencephalogram data. They show that a tipping point marks the brain’s nonlinear wake-to-sleep transiti...

My co-authors have yet to move to Bluesky, so I'm pleased to announce our latest work has just been published in @nature.com Neuroscience. Amazing work led by Junheng Li, revealing that falling asleep follows a predictable bifurcation pattern #neuroskyence #sleep
www.nature.com/articles/s41...

Screenshot of the article that reads: While parents are asked to consent to these surveillance programs, the school’s handbook says that there is “no expectation of privacy” on campus. If parents want to restrict recordings of their children outside of the school, they have to manually opt out of the “anywhere” option—as Jessica Lopez learned midway through her older daughter’s second year at Alpha. The girl says she remembers sitting on her bed one night working on schoolwork when she received a notification that she’d been flagged for an anti-pattern. She says Alpha’s system sent a video of her in her pajamas, taken from the computer’s webcam, that showed her talking to her younger sister.

Just monstrously bad

Screenshot of article that reads: Neil Selwyn, an education professor at Monash University and author of Should Robots Replace Teachers?: AI and the Future of Education, says attempts to automate teaching usually underestimate how much the profession requires improvisation and adapting to a particular student’s needs. Alpha’s trust in software-enabled repetition and students’ self-motivation is often typical of education ventures started by people with backgrounds in tech who were self-taught and “then fix onto self-regulated learning or one-to-one tutoring as the way that one can learn math or science or engineering or coding most effectively,” Selwyn says. “But they didn’t tend to learn history, poetry, or archaeology, or any of the humanities” that way.

Assuming that everyone learns the way you do, and then structuring your pedagogy on that premise, it's one of the worst things you can do when working with any students

When #HHMIJanelia released the #Drosophila hemibrain in 2020, rendering all cell types at once with full shading/shadows was too hard, so the video showed them region by region. Technology has since advanced, and I went back and rendered them all together. I like how it shows the internal structure.

The question isn’t "why does Signal use AWS?" It’s to look at the infrastructural requirements of any global, real-time, mass comms platform and ask how it is that we got to a place where there’s no realistic alternative to AWS and the other hyperscalers. 3/

> “Universities are not tech companies,” Olivia Guest and Iris van Rooij, two computational cognitive scientists at Radboud University in the Netherlands[.] “Our role is to foster critical thinking,” the researchers said, “not to follow industry trends uncritically.”

4.1 Cookbookery and Mathematistry The maladies which result may be called cookbookery and mathematistry. The symptoms of the former are a tendency to force all problems into the molds of one or two routine techniques, insufficient thought being given to the real objectives of the investigation or to the relevance of the assumptions implied by the imposed methods.

Not to be that person but from the original Box (1976)

excellent thank you. Of course and to my shame I had never read this.

New preprint! What happens in the brain when people offload memories into external reminders? Using fMRI decoding, we found that the corresponding neural trace fades until it becomes statistically absent.

osf.io/preprints/ps...

🧵...

This is v cool.

Do you think people retain any additional 'familiarity' information for the Offload items?

ie a memory of 'I know I can look that up later, as needed'

The home of the O'Donovan family is being used as a polling station in Coolmoyne, Co Tipperary. Around 250 people are eligible to vote at the house and after they cast their ballot, they will be treated to a cup of tea, a slice of apple or rhubarb tart, or a scone | More: rte.ie/b/1540324

SUNDIALS is great. But more options is always a good thing, right?

I usually use Tsit5, but try out a few others if needed.

We also support and ship solvers from SUNDIALS blogs.mathworks.com/matlab/2024/...

Which of those 50 solvers are useful to you and why?

George Box famously said "all models are wrong, some are useful", but what he forgot to add was that usefulness doesn't just depend on the model.

A model is useful *only with respect to a given target problem*

Spiking NN fans - the #SNUFA workshop (Nov 5-6) agenda is finalised and online now. Make sure to register (free) soon. (Note you can register for either day and come to both.)

Agenda: snufa.net/2025/
Registration: www.eventbrite.co.uk/e/snufa-2025...

Thanks to all who voted on abstracts!

🤖🧠🧪

Join the team at Janelia to use #theory, computation, and modeling to tackle bold questions in #biology. Collaborate across scientific disciplines to drive discovery.

Apply by Nov. 3 @ https://janelia.link/theoryfellowprogram

#computationalbiology #sciencecareers

"statisticians" have missed a lot of boats due to staying in lanes, imo

yeah I agree with you that is very frustrating, and you often end up guessing/stressing what things are worth keeping vs cutting. imo the responsibility should be on the funder to manage applicant, reviewer and panel member expectations accordingly.

Or ultimately we should just switch to a lottery!

that's good to hear, but even today when I check the MATLAB page there are 10 solver options listed (uk.mathworks.com/help/matlab/...).

Whereas Julia has all those, plus >50 more: docs.sciml.ai/DiffEqDocs/s...

personally I like the short applications, some schemes are ridiculously long format which is a big time sink for what is usually a low chance of funding. The number of wasted people hours on failed grant applications is criminal.

But ofc reviewers should adjust their expectations accordingly

Ireland has a presidential election this Friday (largely ceremonial position), 2 candidates and the left-leaning independent one Catherine Connolly is well ahead in the polls.

Presumably because of this she is being hit hard this week by newspapers but now disturbingly also deepfake disinformation

Screenshot of text: "We have this massive surplus of nutrients that means it's an enriched broth," is how Dr Adam Mellor from the Agri-Food and Biosciences Institute (AFBI) describes Lough Neagh. Those nutrients are primarily phosphorus - one of the building blocks of life - and they come from many sources. "We have waste water as one of the significant contributors and we have land use and agriculture probably as the other large contributor. "But we also have individual contributions through septic tanks and what we individually put into the wastewater system."

Screenshot of text: "An exercise carried out by AFBI determined that agriculture was responsible for about 60% of the nutrient load in the lough, wastewater approximately a quarter and the rest from septic tanks, industry and households."

odd framing in this (otherwise v good) BBC article on Lough Neagh pollution. Early on quotes Mellor listing 6 sources with agriculture named 3rd

Later describes report from ABFI (Mellor's employer) that says agriculture responsible for 60% of nutrient load in the lake

www.bbc.co.uk/news/article...

For this reason I prefer the UK PhD model of 3-4 years with fairly hard time limit.

Do you learn less and achieve less than in a US PhD? Probably.

Do you still learn *enough* to call yourself an expert in a topic? Usually, yes.

Let people move on with their lives

The biologist's view of how flight works. Courtesy of Rory Maizels. #GenerativeBiology

Researchers applying for the UK's Global Talent visa face high upfront costs Upfront cost of visa to applicant and employer in 2025 UK Denmark £692 India £602 Australia £405 Israel £362 US £305 Italy £235 Netherlands 1 £207 Germany | £170 Spain 1£144 Sweden | £117 South Korea | £98 France £84 Japan £21 £5,941 £O £2,000 £4 Source: Fragomen for the Royal Society • Excludes ongoing costs such as mandatory healthcare premiums or fees such as language tests

The costs of the UK’s Global Talent Visa looks a trifle high in comparison to competitor countries - largely through the Immigration Health Surcharge (which critics say is a form of double taxation as they contribute to the NHS through normal tax on their earnings)